Date of Award


Document Type


Degree Name

Doctor of Philosophy (PhD)



First Advisor

Phillip J. Brantley


The present study used a cross-sectional design to answer the following questions: (a) what is the prevalence of DSM-IV depressive and anxiety disorders in a population of low-income primary care patients with type 2 DM, and does a diagnosis of type 2 Diabetes Mellitus (DM) contribute to an increased prevalence of affective disorders above the rate identified in other chronic illness groups and those suffering from no medical diagnoses, (b) does perceived social support, as measured by the Interpersonal Support Evaluation List (ISEL) have a direct and/or buffering effect on the association between chronic disease and affective disorder diagnoses, and (c) which aspects of social support (appraisal, tangible, belonging and self-esteem) serve as the best predictors of a diagnosis of an affective disorder in patients with chronic illness. The sample included 326 randomly selected adult female patients recruited from primary care clinics at a public hospital. The sample consisted predominantly of uninsured, African American, low-income, middle-aged females. Logistic regression analyses identified a significant main effect for illness group when age and education were statistically controlled (X 2 = 22.66, df 4, p = .000). When posthoc comparisons were examined, significant contrast effects occurred when type 2 DM were compared with other chronic illnesses. Specifically, the odds of having an affective disorder Logistic regression also identified a significant interaction between social support and illness (X2 = 35.42, df 5, p = .000). Results indicated that social support was more beneficial for patients with chronic illness. Each SD decrease in social support increased the odds of having an affective disorder by 67% for the total sample. Although tangible support was identified as an important buffer for affective disorders, emotional sources of support appear to be equally important. Given the high prevalence of affective disorders identified in this sample and the beneficial effect of social support, intervention implications are suggested for those working in primary care settings.